Patent Application
20080220296
The present invention relates to PtRu nanoparticles, catalysts, and a method of using the same in proton exchange membrane (PEM) fuel cells, for example.
In order to improve the performance of PEM fuel cells, as well as reducing costs thereof, it is necessary to reduce overpotentials associated with the O2 reduction reactions (ORR) and to lower cathode precious metal (Pt) catalyst loading with more cost effective architectures. In addition, improvement of tolerance to contaminates, such as CO, is critically important in order to effect of large scale implementation. Pt alloys with enhanced activities have been identified as one of the most promising materials for meeting these challenges. A wide array of studies have addressed conventional Pt alloy electrocatalysts and, more recently, nano-architectured electrocatalysts for improved O2 reduction. While significant progress has been made at reducing cathode Pt loading in PEM fuel cells, cathode electrocatalyst compositions and architectures that provide improved performance for reduced loading down to 0.2 mg of Pt/cm2, for example, and long-term stability under electrochemically active conditions have not yet been identified.
The development of bimetallic heterogeneous catalysts has historically been achieved mainly through chemical intuition and empirical synthetic approaches. Catalytic reforming, fuel cell electrocatalysis, hydrodesulfurization and partial alkene oxidation are a few examples of important technologies that rely on bimetallic systems which have been developed over the last several decades. Recent advances in surface science techniques, analytical instrumentation and first-principles calculations provide some mechanistic insight into the atomistic surface chemistry governing catalytic activity and offer a basis for true rational design of heterogeneous catalysts. However, to develop bulk scale catalysts beyond fundamental surface science studies, it is necessary to develop and couple new nanoparticle (NP) synthesis methods with first principles of theoretical design and surface science modeling-studies. To date, this has not been accomplished for PtRu bimetallic nanoparticles, which are producible in bulk in a controllable manner, and which exhibit improved catalytic performance for reduced loading as well as long term stability under electrochemically active conditions.
Thus, a need exists for PtRu bimetallic nanoparticles which can be produced in bulk in a controllable manner, and which exhibit improved catalytic performance for reduced loading as well as long term stability under electrochemically active conditions.
Accordingly, it is an object of the present invention to provide PtRu core-shell nanoparticles (NPs) which exhibit a new activity and selectivity.
It is, further, an object of the present invention to provide PtRu core-shell nanoparticles having improved catalytic performance for reduced loading.
Moreover, it is an object of the present invention to provide PtRu core-shell nanoparticles having long term stability under electrochemically active conditions.
It is also an object of the present invention to provide a proton exchange membrane fuel cell containing the present PtRu core-shell nanoparticles as catalysts, as well as a vehicle powered by the fuel cell.
The present invention is based, in part, upon the discovery of a nanoparticle (NP) catalyst containing a Ru core covered with a shell of Pt atoms (i.e. a Ru@Pt core-shell NP) that has predictable catalytic properties that are markedly different from nanoparticles of “bulk” PtRu alloys or monometallic Pt and Ru mixtures of identical loadings and compositions. The present inventors demonstrate herein the unique properties of the Ru@Pt NP catalyst by way of preferential CO oxidation in hydrogen feeds (PROX), a reaction of key importance for the practical implementation of hydrogen fuel cells. Our DFT studies have indicated the origin of the enhanced activity for the core-shell NP's and provide a fundamental mechanistic explanation of the hydrogen-promoted CO oxidation reaction at low temperatures. The results disclosed herein show that the electronic structure, catalytic activity and chemical selectivity of bimetallic heterogeneous catalysts can now be designed, implemented and turned through a combination of theoretical analysis and core-shell NP synthesis in a controllable and reproducible manner.
We have shown that Pt monolayers on base metals sustain the high activity of pure Pt for H2 activation kinetics, whereas at the same time bind adsorbates (e.g. CO) much weaker than pure Pt. The core-shell architecture provided by the present invention, where only one type of atom is present on the surface, invokes a combination of “ligand” and surface strain effects without any mechanistic complications of the alloy surface bifunctionality. The latter refers to more common bimetallic catalysts, where both alloy components are present on the surface. In such bimetallic systems, the more oxophilic metal acts as an oxygen activator (i.e. to form surface OH), which facilitates the oxidation of the CO adsorbed on neighboring, less oxophilic metal centers. In contrast, core-shell catalysts have only one type of surface metal but their electronic structure and catalytic properties are substantially modified because of the interactions of the shell atoms with the core atoms. The kinetically-stabilized core-shell structure has already proven itself as a novel architecture for NO reduction over the PtCu bimetallic system. In situ chemical and electrochemical deposition of Pt and other metals onto core NPs (including Ru) has been reported but the resulting catalyst NP structure/architecture is difficult to assess in these systems. However, the lack of controlled bulk synthetic procedures and the limited structural and spectroscopic information for PtRu systems has hindered mechanistic interpretation and direct comparison with other NP architectures.
The present invention is based, at least in part, upon the discovery of a Ru@Pt core-shell NP catalyst that is distinctly different from PtRu alloy and from the mixed monometallic systems of the same composition. In PROX reactions, the Ru@Pt NPs are far more active for CO oxidation than alloy and monometallic NP catalysts. Because the Ru metal is confined and kinetically trapped inside a Pt shell, the conventional bifunctional mechanism cannot be implicated since CO oxidation necessarily occurs entirely on the Pt surface sites; no Ru is exposed on the NP's surface. Through DFT modeling, the present inventors have determined that the enhanced CO oxidation is achieved through modification of the electronic structure of the Pt surface by the presence of subsurface Ru. This modification significantly destabilizes CO on Pt, leading to lower CO saturation coverage, thereby providing more adsorbate-free active sites where O2 and H2 can be activated. At the same time, this electronic modification greatly accelerates the CO oxidation reaction through a substantial destabilization of the adsorbed reactive intermediates.
The present invention is also based, at least in part, upon the discovery of a synthesis of a nanoparticle (NP) catalyst containing a Ru core covered with an approximate 1-2 monolayer-thick shell of Pt atoms (i.e. a Ru@Pt core-shell NP). The distinct catalytic properties of these well-characterized core-shell nanoparticles have been demonstrated for preferential CO oxidation in hydrogen feeds (PROX) and subsequent hydrogen light-off. For H2 streams containing 1000 ppm CO, H2 light-off is complete by 30° C., which is significantly better than traditional PtRu nano-alloys (85° C.), monometallic mixtures of nanoparticles (93° C.) and pure Pt particles (170° C.). Density Functional Theory (DFT) studies suggest that the enhanced catalytic activity for the core-shell NP's originates from a combination of (i) an increased availability of CO-free Pt surface sites on the Ru@Pt NP's, which are necessary for O2 and H2 activation, and (ii) a hydrogen-mediated low-temperature CO oxidation process that is clearly distinct from the traditional bi-functional CO oxidation mechanism. These characteristics of the PtRu core-shell nanoparticles disclosed herein are plausibly considered to be responsible for the observed properties thereof. These properties are:
1) improved catalytic performance under reduced Pt loading, and
2) long term stability under electrochemically active conditions.
Generally, the nanoparticles of the present invention have a size of from about 1-15 nm, and preferably from about 3-10 nm. Most preferably, the nanoparticles have a size of from about 4-8 nm. These Ru@PT NPs exhibit a superior combined activity and selectivity for oxidizing hydrogen in the presence of CO in CO-rich gas feeds having from about 10 to 10,000 ppm of CO therein.
The following terms used throughout the present specification and claims are as defined herein below:
1) Pt and Ru are the metals platinum and ruthenium, respectively. The nanoparticles described herein may be referred to interchangeably as PtRu nanoparticles, Ru@Pt NPs, Ru@Pt core-shell NPs or PtRu core-shell nanoparticles.
2) nanoparticle generally means particles having a size of less than 100 nm, and as also particularly defined throughout this specification.
3) TEM means Transmission Electron Microscope.
4) XEDS means X-ray Energy Dispersive Spectroscopy.
5) PVP means polyvinylpyrrolidone, and of a molecular weight in the range of about 30,000 to 75,000, or as specifically indicated in the present specification.
6) high-boiling solvent means a solvent having a reflux temperature of at least 150° C., and preferably at least 175° C.
Thus, the present invention provides a nanoparticle architecture containing a Pt shell that coats a ruthenium/ruthenium oxide nanoparticle core. The structure of the particle is different from traditional alloy particles and it is believed to be the architecture and the chemical state of the metals in the nanoparticles that provides the superior catalytic activity.
The nanoparticles of the present invention are more efficient at oxidizing hydrogen in CO-rich gas feeds up to 1000 ppm of CO than any other catalysts reported to date. Thus, the present nanoparticles provide a superior anode catalyst for hydrogen fuel cells that is far more active and tolerant of CO impurities than are the current state-of-the-art catalysts. The present nanoparticles are also more efficient at activating molecular oxygen for ORR processes. In addition, the Pt shell thereof prevents acid attack of the Ru core in PEM fuel cell conditions. Hence, the present nanoparticles maybe acid-resistant under electrochemically active conditions.
Generally, any platinum salt precursors that can be reduced to the metallic state, such as PtCl2 and Pt(acac)2 [acac=acetylacetonate], H2Pt (IV) Cl6, PtCl4, or any Pt complex that can be reduced to the metallic state, such as Pt (C2H4)3, Pt (COD)2, Pt (PPh3)4, may be used as Pt-shell material. Any ruthenium salt precursor that can be reduced to the metallic state, such as RuCl3, Ru(acac)3, Ru2(CO)6Cl4 and mixtures thereof, can be used as a starting material for the synthesis of Ru-core.
Further, in general, any polyols with boiling points in excess of 150° C., such as propylene glycol (propane 1,2-diol), trimethylene glycol (propane 1,3-diol, diethyhlene glycol, triethyelene glycol etc. may be used as solvent. Any organic solvent inert to ruthenium (catalyst) can be used with a suitable reducing agent and ligand combination.
Examples of reducing agents are alcohols, amines, NaBH4, butyllithium, methyllithium, hydrazine or other similar agents. Examples of other solvents are decahydronaphthalene, octylethers, oleylamine, hexadecane, trioctylphosphine, diglyme, glycol, hexanediol or combinations thereof.
Powder x-ray diffraction technique is used to characterize the Ru@Pt core/shell NPs. The particles might also be made amorphous, however, and, therefore, give no diffraction. Notably, the architecture (i.e. Ru core with a Pt shell) is an important feature but the structure (Ru metal and Pt metal as shown by XRD) is not limative.
The present NPs as made may have anywhere from 100% Ru (0)-0% (Ru+4) to 0% Ru(0) to 100% (Ru+4) See also the examples below.
The synthesis of the present Ru@Pt core-shell NPs is conducted using Schlenk line techniques. Such techniques as well-known and generally used for the preparation of air-sensitive compounds.
TEM imaging reveals small fairly monodisperse nanoparticles of about 3.0 nm mean size. See
Having described the present invention, reference will now be made to certain examples which are provided solely for purposes of illustration and are not to be considered limitative. All preparations of the present Ru@Pt core-shell NPs below were conducted using Schlenk line methodologies.
Pt is coated over as-made Ru seeds in a separate deposition sequence. 54 mg PtCl2 is dissolved in 40 mL colloidal Ru suspension, the mixture is heated to 130° C. under vigorous stirring, and then brought to a boil with a temperature ramping as slow as 1-2° C. per minute. The reaction is quenched by removing the reaction flask off the mantle after 1.5 hours of constant refluxing. The Ru@Pt nanoparticles show a mean particle size greater than that of monometallic Ru nanoparticles, with a narrow size distribution (
Randomly chosen NPs for STEM point analysis have both Pt and Ru elements in 25-75% range (
Micro-Raman spectrum exhibits a similar pattern as monometallic Ru NPs, except a shoulder at 590 cm−1. Annealing the particles at 700° C. for 2 hours yields a sharper feature centered at about cm−1 along with all modes for the rutile phase Raman shifted to higher frequencies as shown in
PtRu alloy NPs may be synthesized by co-deposition in high boiling solvent, as noted above. Generally, as noted above, and as another example of the present invention, metal precursors of Ru and Pt, such as [Ru(CO)3Cl2]2 dimer and Pt(acac)2, for example, are co-reduced in glycol while being capped with PVP. Generally, PVP molecular weights of 30,000 to 75,000 are used, while molecular weights from about 40,000 to 60,000 are preferred. A typical synthesis may use [Ru(CO)3Cl2]2, Pt(acac)2 and 55 mg PVP55000 in glycol. X-Ray diffraction pattern in
Catalysts may be prepared by adding a support material, such as alumina, preferably γ-Al2O3, to a colloidal suspension of nanoparticles, and drying the slurry under vacuum. Typically, a suspension of nanoparticles and γ-Al2O3, for example, are mixed and vacuum dried at temperatures over 100° C. while vigorously stirring the mixture. Such composition yields a 1% by weight Pt alumina supported bimetallic catalyst. The catalyst is washed with polar organic solvent, such as acetone several times and equip-volume mixture of acetone and ethanol, for example, then baked at 60° C. overnight.
The catalysis is carried out using 105 mg catalyst of any kind. The catalytic rig is designed in a flow-through fashion. An inlet velocity of gases of 0.21 m/s, and a total flow rate of 400 NmL/min is employed. The gas mixture for the PROX reaction is composed of 0.1-0.2% (99.999% pure). The catalysts are reduced in 50% H2 (99.999% pure), and balance Ar (99.999% pure). The catalysts are reduced in 50H2 at 200° C. prior to catalysis. The temperature is set to 200° C. and the heating ramp is 1.6° C./min. The rate of H2O formation is plotted versus temperature for each core@shell, alloy and physical mixture NPs catalysts for 0.1% and 0.2% by volume CO contaminated H2 feeds, in
The Ru@Pt core-shell NPs were synthesized by using a sequential polyol process. Ru(acac)3 (acac=acetylacetonate) was initially reduced in refluxing glycol in the presence of PVP stabilizers (MW=55,000). The resulting Ru NPs (mean particle size=3.0 nm) were subsequently coated with Pt by adding PtCl2 to the Ru/glycol colloid and slowly heating to 200° C. The PtRu alloy NPs were synthesized via co-reduction of the [Ru(CO)3Cl2]2 dimer and Pt(acac)2 with glycol and PVP stabilizer at 200° C. Monometallic Pt NPs and Ru NPs were prepared from PtCl2 and Ru(acac)3, respectively, using slight modifications of published procedures. To make a physical mixture of monometallic Pt and Ru NPs, the separate colloids were mixed. All catalysts were prepared with 1.0 wt % Pt loadings by impregnating γ-Al2O3 supports with the colloids in accordance with a known procedure.
The Ru@Pt NPs show a mean particle size of 4.1 nm (
The XRD profiles of the Ru@Pt NPs show face centered cubic (FCC) diffraction peaks for the Pt shell with an additional reflection at ˜42° (a shoulder next to the Pt (11) peak at 39.8°) that arises from the poorly-crystalline HCP Ru core. The refined Pt lattice parameter (Lebail profile fit) for the Pt shell of the Ru@Pt particles gives a 3.910(1) Å FCC lattice constant, which is slightly compressed from that of pure Pt at 3.923 Å. While diffraction from monolayer films has been well described in a theoretical framework, it is rarely observed due to the lack of scattering matter from a thin film surface. Bulk samples of monolayer coated NPs provide a higher density of scattering matter and enhanced X-ray diffraction relative to thin film samples. The XRD data for the Ru@Pt particles with approximate monolayer coverage show relatively strong Pt 111 diffraction peak whose peak position is shifted to higher 2θ compared to bulk Pt and is consistent with a compressed lattice. Additionally, the 002 reflection of Ru@Pt is shifted from its normal position to lower 2θ and has a lower intensity relative to bulk Pt. Monometallic Pt NPs synthesized under identical conditions show bulk Pt diffraction patterns with no anomalies in their peak positions. As the Pt shell becomes thicker with additional overlayers, the peak positions for the 111 diffraction shift to their “normal” position with increasing intensities. We attribute the anomalies in the diffraction data for the monolayer shells to incomplete lattice formation and strains associated with the 2D structure. These anomalies also suggest that the observed diffraction peaks do not arise from low concentrations of pure Pt NPs in the Ru@Pt sample.
Annealing the Ru@Pt NPs at 500° C. in vacuum induces alloy formation, which is evidenced from the contracted FCC unit cell with a=3.889(1) Å (
X-Ray photoelectron spectra (XPS) for the Ru@Pt NPs show a Ru:Pt ratio of 58:42, which is again consistent with the precursor composition. The Pt 4f levels show two signals; one for Pt metal (80%) and another for PtO (20%), that are both shifted to higher binding energies relative to the monometallic Pt NPs and the Pt standard. The Ru XPS data show metallic Ru (67%) and R4+ (33%) components. The latter is attributed to RuO2. For Ru, the metal and Ru4+ levels are shifted to lower energy relative to the monometallic Ru NPs. Similarly, the PtRu alloy NPs show a total of 45% Pt by atom, about 80% of which is for Pt metal. The Ruo:R4+ ratio is 63:37. However, other Ruo/Ru+4 ratios may be obtained, and even 100% Ru0. The electronic changes in the core levels of Pt and Ru atoms, as determined by XPS, are also consistent with alloy formation. Chloride was not detected in any of the samples.
To probe the NP surface composition, the as-prepared NPs were dosed with CO in the colloidal suspension and subsequently monitored by FT-IR (
The combined TEM, XRD, XPS and IR-CO probe data are all consistent with the core-shell structure for the Ru@Pt NPs and clearly differentiate them from the PtRu alloy NPs. The XRD and XPS studies all suggest an amorphous mixed Ru+4/Ru0 core that is coated by a Pt shell. Pure Ru NPs show the same characteristics except for higher Ru+4/Ru0 ratios and slightly higher binding energies. Empirically, we observe that the RuO2 shell is required for Pt coating. Ru particles prepared under rigorous anaerobic conditions do not provide good seeds for core-shell particle growth and result in the formation of phase-separated monometallic mixtures. However, Ru NPs are readily reduced to the metallic state in flowing H2 at room temperature. As such, it appears that the catalytically active Ru@Pt NPs are metallic after conditioning in H2 and these structures have significantly different activities from those of the PtRu alloy and monometallic Pt and Ru structures under identical loadings and conditions.
To compare and contrast the activity of the core-shell NPs with that of the alloys and monometallic NPs, we evaluated the PROX reaction using H2 feeds contaminated by 0.1-0.2% CO by volume, along with 0.5% O2. The temperature programmed reaction (TPR) data for the core-shell, alloy and monometallic mixture are shown in
The Ru@Pt core-shell catalysts show the highest activity for all the different architectures studied to date (
The Ru@Pt catalyst can be cycled at 200° C. for several hours without loss of activity. Importantly, annealing the Ru@Pt catalyst at 500° C. for 12 hrs induces alloy formation (
The present Ru@Pt NPs may be used advantageously in many reactions, such as PROX, ORR, WGS, reforming and those generally involving hydrogenation and CO-tolerant electrocatalysis.
Furthermore, since the present Ru@Pt NPs exhibit a superior combined activity and selectivity for oxidizing CO in the presence of hydrogen in CO-rich gas feeds, as defined above, these NPs may be used advantageously as anode catalysts for PEM fuel cells using a CO-rich hydrogen feed.
Examples of CO-rich hydrogen sources include off-gases from various industrial processes.
Also, examples of PEM fuel cells may be noted from U.S. Pat. Nos. 6,020,083; 6,010,798 and 7,108,937, which are all incorporated by reference herein in the entirety.
U.S. Ser. No. 11/638,572, filed on Dec. 14, 2006 and U.S. Ser. No. 60/883,845, filed on Jan. 8, 2007, are both incorporated by reference herein in the entirety.
Having described the present invention, it will be apparent to one of ordinary skill in the art that many changes and modifications may be made to the above-described embodiments with departing from the spirit and the scope of the present invention.
This application claims priority to U.S. Ser. No. 60/883,845, filed on Jan. 8, 2007, in the U.S. Patent and Trademark Office.
The work leading up to the present invention was sponsored, at least in part, by NSF CHE 0401850. As such, the U.S. government may have certain rights in the present invention.
| Number | Date | Country | |
|---|---|---|---|
| 60883845 | Jan 2007 | US |
